Children's working hours, school enrolment and human capital ...

3 downloads 0 Views 291KB Size Report
1. Children's working hours, school enrolment and human capital accumulation: evidence from. Pakistan and Nicaragua. F. C. Rosati. M. Rossi. October 2001.
Understanding Children’s Work Project Working Paper Series, October 2001

1.

Children’s working hours, school enrolment and human capital accumulation: evidence from Pakistan and Nicaragua

F. C. Rosati M. Rossi October 2001

Children’s working hours, school enrolment and human capital accumulation: evidence from Pakistan and Nicaragua F.C. Rosati* M. Rossi*

Working Paper October 2001

Understanding Children’s Work (UCW) Project University of Rome “Tor Vergata” Faculty of Economics V. Columbia 2 00133 Rome Tor Vergata Tel: +39 06.7259.5618 Fax: +39 06.2020.687 Email: [email protected]

As part of broader efforts toward durable solutions to child labor, the International Labour Organization (ILO), the United Nations Children’s Fund (UNICEF), and the World Bank initiated the interagency Understanding Children’s Work (UCW) project in December 2000. The project is guided by the Oslo Agenda for Action, which laid out the priorities for the international community in the fight against child labor. Through a variety of data collection, research, and assessment activities, the UCW project is broadly directed toward improving understanding of child labor, its causes and effects, how it can be measured, and effective policies for addressing it. For further information, see the project website at www.ucw-project.org.

This paper is part of the research carried out within UCW (Understanding Children's Work), a joint ILO, World Bank and UNICEF project. The views expressed here are those of the authors' and should not be attributed to the ILO, the World Bank, UNICEF or any of these agencies’ member countries.

*

UCW-Project and University of Rome “Tor Vergata”

Children’s working hours, school enrolment and human capital accumulation: evidence from Pakistan and Nicaragua

Working Paper October 2001

ABSTRACT We analyse the determinants of school attendance and hours worked by children in Pakistan and Nicaragua. On the basis of a theoretical model of children’s labour supply, we simultaneously estimate the school attendance decision and the hours worked by Full Model Maximum Likelihood. We analyse the marginal effects of explanatory variables conditioning on the “latent” status of children in terms of schooling and work. We show that these effects are rather different, and discuss the policy implication of this finding. Finally, we use our predicted hours of work to analyse the effects of work on children’s school achievements.

Children’s working hours, school enrolment and human capital accumulation: evidence from Pakistan and Nicaragua

Working Paper October 2001

CONTENTS

  1.

Introduction ........................................................................................................................... 1

2.

A theoretical outline .............................................................................................................. 3

3.

The econometric model ....................................................................................................... 6

4.

The data sets ........................................................................................................................... 9 4.1 Pakistan ............................................................................................................................. 9

5.

4.2 Nicaragua ........................................................................................................................ 13 Estimates of children’s labour supply. ............................................................................. 15

6.

Impact of hours worked on school performance .......................................................... 19 6.1 Pakistan ........................................................................................................................... 19

7.

6.2 Nicaragua ........................................................................................................................ 21 Conclusions .......................................................................................................................... 23

References .......................................................................................................................................... 32 

1

UCW WORKING PAPER SERIES, OCTOBER 2001

1. INTRODUCTION Child labour is thought to be harmful in many ways to children’s welfare. It interferes with human capital accumulation and may affect the present and future health of the child. The determinants of child labour supply have been recently analysed in the literature (see Basu (1999), Rosati- Tzannatos (2000), Cigno – Rosati (2001) and the literature therein cited for the discussion of theoretical models and empirical results). The attention of the literature has been mainly focused on the determinants of the categorical decision of the household on the activity of the child: whether to send a child to school, to work or allow him to perform both activities1. Almost no attention has been paid to the amount of time that children devote to work (either when this is their only activity or when they combine it with school attendance). The number of hours spent working is not only important in itself as a measure of child welfare (this is a measure of forgone leisure, etc.), but is also an essential ingredient to evaluate the cost of work in terms of health and human capital accumulation. In this paper we analyse the hours of work supplied by children and try to identify their effects on school achievement. Schooling is obviously one of the most important inputs in human capital accumulation, but education is also important for health. Many studies, starting from the seminal paper of Grossmann (1972, 1982), have shown that education has an important positive effect on health status. Hence, lower school achievement due to work may not only influence the human capital of the child, but also its future health. The literature on child labour has mainly focused on the participation decision of the children. Almost no attention has been given to the hours supplied. An exception is Ray (2000), which, however, treats labour supply separately from the household decision of sending a child to school. This paper innovates on the existing literature by focusing on the simultaneous decision relative to school attendance and to the amount of work supplied. On the basis of a simple theoretical model, we estimate a simultaneous two equations system. This model not only allows us to take into proper consideration the joint decisions about work and schooling, but also allows us to calculate marginal effects conditioning on the “latent” propensity of the child to attend school and/or to work. These marginal effects are in some cases rather different across the “latent” state of the child and this has interesting analytical and policy implication. 1

For the quantitatively non negligible cases in which children appears to neither work nor go to school se the literature cited.



CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

The final section of the paper is devoted to an exploration of the link between child work and school achievements. Unfortunately, as it is also true for the rest of the small literature on the subject, the available data sets offer proxies of school achievements that are not fully satisfactory. However, as discussed in more details below, we innovate on the existing literature by taking into account the endogeneity of the hours worked and by disaggregating the effects of work on school achievements by sector of activity. A few attempts have been made to evaluate the impact of child labour on school achievement (Patrinos and Psacharopulos (1994), Psacharopulos (1997), Heady (2000), but their estimates do not take into account the simultaneity of the decision on whether to send a child to school and/or to work. Moreover, they discuss mainly the effects of work on school achievement by considering whether the child is working or not, without devoting much attention to the number of hours actually worked. While these studies offer some evidence on the matter, it is important to assess not only the effect on school outcome of the fact that a child is working, but also the effect of the amount of work performed. From a policy perspective it is important to evaluate whether it is the fact that a child works or if it is the amount of time that she works that affects human capital accumulation. If working hours had only a negligible effect on school achievement, then school attendance rather than work would be the correct policy target (at least in terms of human capital accumulation). On the other hand, if working hours strongly affect human capital accumulation, then child labour also needs to be monitored. Moreover, establishing whether a few hours of work have a negligible effect on human capital accumulation will help to define the target of policy intervention and also offer some supporting or contradictory evidence to the claim often made that children working less than two or three hours a day should not be considered as “working” children.

3

UCW WORKING PAPER SERIES, OCTOBER 2001

2. A THEORETICAL OUTLINE To outline our theoretical model we consider an altruistic set up, where parents care about the present and future consumption and current leisure of their children 2. The number of children is taken as given and for simplicity of exposition is normalized to 13. We also assume that human capital accumulation is the only way to transfer resources for children’s future consumption4. Human capital is accumulated by sending children to school5. The time a child has to spend at school is fixed at hS. Normally school hours are not flexible and school attendance requires a minimum fixed amount of time devoted to school. Some of the children that work and attend school might miss classes and thus make their school hours more “flexible”. However, the degree of “flexibility” that can be achieved in this way is rather limited, as skipping school often results in dropping out and is normally not tolerated by school authorities6. Hence we treat school hours as fixed. School attendance does not rule out child labour. However, we assume that working hours have a negative influence on human capital accumulation. Hours spent at work reduce time available for study, tire the child and reduce her learning productivity, etc. Given the nature of the work that children perform, mainly unskilled and mostly at their family farm or business, we can safely consider the hours spent at work, hL, as flexible and treat them as a continuous choice variable. The human capital production function takes the form:

2

As discussed in Cigno-Rosati (2000), similar results will be obtained if a non-altruistic model were used.

3

Endogenous fertility does make a difference to child labour analysis (See Rosati-Tzannatos 2000), but for the present analysis nothing of substance is changed by treating fertility as exogenous.

4

If capital market were present the efficient level of human capital investment will equalize returns to human capital investment to the market interest rate. Allowing for the presence of capital markets will complicate the exposition without bringing additional insights. For a discussion of the role of capital markets in determining child labour supply see RosatiTzannatos (2000).

5

Child labour could also contribute to human capital accumulation by, for example, on the job training. We do not consider this case in our discussion for two reasons. Firstly, there is no evidence to substantiate the statement on the role of child labor as a means to accumulate human capital. Secondly, formal education plays an empowermet role that goes beyond that of increasing the productivity of working time. This effect is captured in our model by introducing human capital as such as an argument of the utility function.

6

There are programs that try to make school hours more flexible to accommodate child labour activities, but their coverage is marginal and, in any case, such programs are not present in Pakistan.

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA



H=h (hL; hS)

(1)

Where ∂H / ∂hL < 0 . Parents maximize a utility function defined over the current consumption of the household members, the current leisure and the future consumption of the children. Current household consumption C1 is given by: C1S=y +w hL – q

(2)

if parents send their children to school. Where y is the (exogenous) income of the parents, w is the wage rate (marginal product) of child labour, hL are the hours of work supplied by children and q is the direct cost of education. Future children’s consumption, C2S, is given by K+H where K is the exogenous endowment of human capital and H is defined in (1). Parents also attach value to the (current) leisure enjoyed by the children, L= 1 - hS hL (having normalised total available time to 1). If parents do not send their children to school, present consumption is given by C1L= y + w hL, future consumption by C2L = K and current leisure by L= 1- hL. In both cases the choice variable is hL (the time spent at work), but the money and time budget constraints are different according to whether the child is sent to school or not. As the amount of time required by school attendance is fixed, the parent’s choice of hL is given by Max [US* (hL),UL* (hL)] where

US * = max U(y+ w h L - q, K + H(hL ; h S ),1 - h S - h L ; M) hL

(3)

and

U L * = max U(y + w h L , K, 1 - h L ; M) hL

(4)

5

UCW WORKING PAPER SERIES, OCTOBER 2001

and M represents a vector of household characteristics like education of the parents, locality of residence, etc. In other words, parents compare the maximized utility under the two regimes and select the one that yields the highest welfare. Children’s hours of work are hence chosen conditional on the decision whether to send children to school or not. The comparative statics properties of the model show that an increase in parent’s income increases the probability that a child attends school and reduces the numbers of hours worked. An increase in the cost of schooling reduces human capital accumulation. These results, however, depend on the simplifying assumption of exogenous fertility and absence of capital markets. Relaxing such assumptions would not change the results relevant to the focus of the present paper, but it will make a difference for the discussion of child labour policies. A detailed analysis of these issues can be found in Rosati- Tzannatos (2000). Note that child labour supply is expected, other things being equal, to be lower when children are attending school, because of the cost work has in term of human capital accumulation and the higher marginal value of leisure. Also observe that corner solutions are possible in both regimes for hL.

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA



3. THE ECONOMETRIC MODEL As illustrated in the Section 2, the decision of schooling and working are simultaneous. In particular we observe that a child is enrolled in school if Us* - UL*>0 ⇒ s* > 0 and that the hours of work supplied by the children depend also on their enrolment status. We model hours worked and enrolment status starting from the following structure:

s* = Z ' g + u

(5)

h* = X ' b + γs * +ε = X ' b + γ (Z ' g + u ) + ε

(6)

By using the following useful definitions: X*’b*=X’b+γZ’g and η= γu + e, we can rewrite equation (6) as follows:

h* = X * β * +η (6)

h* are the hours worked, s* is the enrolment status of the child,

ε and u are the disturbance terms following a bivariate normal distribution with zero means and variance co-variance matrix (Σ) as follows:

⎡σ ε2 ⎢ σ = ⎣⎢ εu

σ εu ⎤

⎥ 1⎦ .

We allow the two equations to be correlated via their error terms. One possible source of correlation is the unobservable (by the researcher) ability of the child. If children with higher abilities are more likely to go to school and work fewer hours, we expect a negative correlation between the two error components. Both the enrolment rate and the hours worked are latent variables. Enrolment is observed as a dichotomous variable according to the following structure:

7

UCW WORKING PAPER SERIES, OCTOBER 2001

s=1 if s*>0 s=0 if s*0 h=0 if h*0, enrolled Working hours=0, enrolled Working hours>0,not enrolled Working hours =0, not enrolled The probability associated to each of the regimes can be written as follows: Pr (1)=P(s=1)*P (h*>0|s=1)

(

⎛ Z ' g + ρσ −1 ( h * − X *' b*) ⎜ 1− ρ 2 ⎝

φ (h * − X *' b*, σ )Θ⎜ =

) ⎞⎟ ⎟ ⎠

Pr (2)=P(s=1)*P (h*0||h*>0)= Φ 2 ( X *' β *, Z ' g , ρ ) / Φ ( X *' β *)

( )

( )

18 

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

presented should be interpreted as relating respectively to children with “low propensity” to work and with “high propensity” to work. Some of the explanatory variables have quite different effects on the two groups. Consider for example income. Children with a high propensity to work show an enrolment sensitivity with respect to income far smaller than that of the children that have a lower “propensity” to work. Analogously, household size has a negative and small effect on the probability of attending school for the potentially working children, while it has a strong and significant positive effect on the other group. As we control for income, this is likely to be a marginal productivity effect. In households that are not likely to send their children to work, substitutability between adult and child work appears to be stronger than in the other group. The presence of younger children reduces the enrolment probability for those children who do not work, while it increases it for those children who are working. These results indicate that policies aiming at reducing child labour by introducing incentive schemes (like income transfers) that only marginally modify the opportunity set of the household are likely to produce significant effects only on those households that are at the margin between sending their children to work or to school, i.e. that have a low propensity to child labour. Such schemes are hence likely, if not properly targeted, to be ineffective toward those households, most likely the poorest and most uneducated, that hence have a high propensity to send their children to work.

19

6.

UCW WORKING PAPER SERIES, OCTOBER 2001

IMPACT OF HOURS PERFORMANCE

WORKED

ON

SCHOOL

6.1 Pakistan To what extent is child labour affecting school performance? The ILOIPEC survey does not have a direct measure of school achievements. We have hence constructed an indicator of falling back in the course of study. In particular, we have utilized a dummy variable taking value of 1 if a child aged 10 or more is still enrolled in the primary school or if aged 13 or more and still enrolled in the middle school, zero otherwise. Obviously this variable is only an approximation to the measure of school achievement, because it does not measure the extent to which a child fell back in his/her course of study nor other dimensions of school achievements. The probability of falling back in the course of study was regressed on the hours worked predicted by the model described above and on a set of household characteristics like income, household composition, area of residence and education of the parents. We have also used different specifications in order to highlight the role played by the sector of work of the child and possible non-linearity in the effects of hours on school achievements. The results are shown in Table 9. Before discussing the effect of the hours of work, let us describe the role played by the other explanatory variables. The education of both the father and the mother reduces the probability of falling back in the course of study, even if the education of the mother is only marginally significant. This confirms the role played by maternal education as an input in the production of human capital (see Behrman et al. 1999). Father education might proxy an income effect, but also might reflect the role of father time as an input in the human capital accumulation. Income has also a negative effect, pointing to the importance of inputs other than time in the human capital production function. Being female increases the chances of falling back in the course of study. As we have no information on the time spent on household chores, the female dummy might be capturing the effect that girls are more likely than boys to spend time in that activity. The increase in the hours worked significantly affects human capital accumulation. The results in column 1 show that an additional hour of work a day increases the probability of falling behind by just over 1.6 percentage point at the mean. As Fig. 1 shows the increment is rather steady at all the levels of hours, even if it tends to become smaller the higher the number of hours worked. We have disaggregated the hours worked by sector of activity of the child. After some preliminary testing on the equality of

20 

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

coefficients, we selected the specification presented in column 2 and Fig 2. The cost in terms of human capital accumulation is higher if the child is employed in trade or manufacturing, smaller if employed in the other sectors (largely dominated by agricultural employment). In the former an extra hour of work increase the probability of falling back in the course of study by just over 2 points (at the mean), in the latter of about 1.5 points. Again the equation shows that increases are higher, even if marginally, for the initial hours of work rather than for the subsequent ones. Overall the estimates show that the effects on human capital accumulation are not linked to the mere fact that a child is working but also depends heavily on the number of hours that s/he is working. As Figures 1 and 2 show, for a child working 3 hours the probability of falling behind in the course of study is 5 points higher than for a child working one hour. The estimates also indicate that the marginal cost in terms of human capital accumulation does not increase with the numbers of hours worked. In fact, if anything the estimates show the opposite: (marginally) decreasing marginal costs. As this result is important in terms of policy (it indicates that also a few hours of work can non trivially influence school outcomes) we have tried to further test for non-linearity in the effects of working hours on school achievements. We have employed various specifications of the hours in the estimates: quadratic, cubic, etc. all the estimates suggest that non linearities are present and that the marginal impact of hours is higher at low levels of hours worked. They also suggest that marginal effects increase when the number of hours worked becomes high. Unfortunately the estimates, even is suggestive of the result just described, are not very precise, mainly because of the difficulty in estimating coefficients with highly correlated variables. An example is reported in column 3 of table 9 where the regressors have been included, beside total hours, as two separate variables the hours worked respectively between 2.4 and 4 hours and above 4 hours12. The estimates show that up to 2.5 hours, one additional hour of work increases of about 3.8 points the probability of failing at school. From 2 to 4 hours the increase in the probability is negligible, and above for hours the increase is about 2.7 point for each additional hour. This is illustrated (after correcting the intercept) in Figure 3. The results obtained show that the marginal cost in terms of human capital accumulation of the hours of work is not substantially different whether a child works just a few or many hours a day. In fact, there is some evidence that these costs are higher for “initial” hours of work. This indicates that not considering as child workers those children that work just a few hours a day

12

The cut off points have been selected through a grid search procedure by looking at the log likelihood.

21

UCW WORKING PAPER SERIES, OCTOBER 2001

might be misleading, as the cost of these few hours of work is not negligible.

6.2 Nicaragua The survey for Nicaragua contains more information than that on Pakistan on the education of the child. In particular, there is information also on the grade attended the previous year by a child currently at school. On this basis, it has been possible to build an indicator of school performance defined as the difference between the grade the child should be attending (according to his/her age) and the one s/he is actually enrolled in. We have used as dependent variables in the estimates a dummy variable taking the value of one if the child has lost at least one year of school and zero otherwise13. This indicator was then regressed on a set of explanatory variables similar to those used for Pakistan. Since the hours worked are endogenous, we used the predicted values to consistently estimate the impact of hours worked on school performance. Table 10 contains the results of the estimates. An additional hour worked significantly increases the probability of having repeated at least one grade. The higher the quintile of income the child belongs to, the lower is the probability of failing at school. Children with more siblings and children living in rural areas have a higher chance of repeating at least one year. Girls appear to have better school performance than boys. Parents’ education has the expected negative effect on the probability of failing at school. The more the parents are educated the higher is the probability that the child is enrolled in the right grade. Mother’s education affects more markedly than father’s education the child’s performance at school. Figure 4 illustrates the relationship between the probability of failing at school and the hours worked, holding constant the other continuos variables at their means and the dummies variables at their mode. As for Pakistan, even a small number of hours worked affects significantly school performance. One hour of work a day has a strong impact on the school performance of the child increasing by almost 3.6 percentage points the chance of failing at school. The probability of having failed at least one grade increase by 10 points if the child works 3 hours and by more than 20 if s/he works 8 hours. Information on the sector of activity is not available in this data set, therefore we were unable to compare the effects of hours worked in 13

Similar results are obtained using the numbers of years that a child has lost or other indicators based on a transformation of this variable.

22 

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

different sectors as we did for Pakistan. Other specifications were also used to test the non–linear effect of hours worked on school performance by using exponential and quadratic form functions of hours. The related estimates confirm the non linear relation between hours worked and school performance and highlight the larger impact of working hours at low levels of hours worked.

23

UCW WORKING PAPER SERIES, OCTOBER 2001

7. CONCLUSIONS The literature on child labour has to some extent neglected to analyse the determinants of the hours worked by the children. The attention has been mainly devoted to the household decision to send the children to school and/or to work. The duration of the working day is, however, important to assess the impact of work on the human capital accumulation and on the child’s health. Starting from a simple theoretical framework, we have derived and estimated a simultaneous equation system for estimating the household’s decision relative to the school enrolment and to the hours worked by their children. The results show the importance of taking into account the simultaneity of the decision about schooling and hours worked in order to assess the importance and the role played by different explanatory variables. We have seen that the number of hours worked also depends on the (endogenous) decision whether or not to let the child attend school. The effects of the variables on the hours worked then depend also on the change they induce in the probability that a child is sent to school. In fact we have seen, that this latter “indirect” effects often dominates the direct one. Moreover, the structure of the model we have estimated allow us to compute not only the “full” marginal effects described above, but also the marginal effects conditional on the latent variable indicating the “propensity” of the household to send the child to work or not. These marginal effects are very different among the two “groups” and show that policy action can have a different impact depending on whether the child is likely to be sent to work or not. Finally, we have obtained consistent estimate of the effects of hours of work on human capital accumulation by regressing our indicator on school achievement on the hours of work predicted by our model and on other variables. The results of the estimates, based on an indicator of whether the child has fallen back during her course of study, indicates that the amount of hours worked are an important determinant of school achievements beyond the fact that the child participates in economic activities. These effects are far from negligible, as a few hours of work per day increases the probability of falling back in the course of study of about 10 per cent. Hours of work have an effect that is not increasing; if anything the first hours of work have a larger impact on school achievements than the successive ones. This indicates that the assumption often made that a few hours of work only have negligible effects on human capital accumulation is not supported by the evidence, at least in the case of Pakistan and Nicaragua. A different issue is whether the changes induced in school achievements are “large”. This has implications in terms of policy interventions. If the loss due to child work is not large in terms of human capital accumulation, then school

24 

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

enrolment rather than child work itself should be targeted (at least as far as the effect of child labour on human capital accumulation is concerned). More evidence and the use of a larger set of indicators are necessary to deal with such an issue.

25

UCW WORKING PAPER SERIES, OCTOBER 2001

Table 6. Descriptive statistics Variables

Pakistan

Mean

Nicaragua

Standard

Mean

(1)

Weekly hours worked Daily hours worked Weekly hours worked if working

15.72 35.29

(2)

Standard deviation

deviation

(1)

(2)

20.74 0.68

2.11

5.94

2.77

16.59

Dailly hours worked if working Age

10.14

2.780

9.82

2.58

Hh size

8.46

3.54

7.74

2.99

Babies

1.38

1.30

3.13

1.45

Children

3.55

1.48

1.04

1.07

2968.34

2588.09 4779.45

5894.62

HH (net) income HH income Female

0.389

0.487

0.49

0.50

Rural

0.435

0.496

0.55

0.50

Father education: primary

0.267

0.442

0.42

0.49

0.19

0.40

Father education: secondary or more Mother education: primary

0.053

0.225

Mother education: secondary or more

0.44

0.50

0.16

0.37

Fail rate

0.706

0.455

0.79

0.40

Fail rate for enrolled children aged 10 or more

0.667

0.471

0.73

0.44

Number of observations

27512

4278

Note: children aged 5-14 and aged 6-14 are the sample considered for Pakistan and Nicaragua, respectively

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

26 

Table 7. ML estimates of enrolment and hours worked. Pakistan Enrolment

Hours* (a)

Parameter

p value

(b)

Marginal

Marginal

effect|

effect|

working

working

(c) Parameter

p value

not

(d)

Total

Total

marginal

marginal

effect|

effect|

enrolled

enrolled

Age

0.592

0.000

0.335

0.207

23.634

0.000

-1.273

-3.454

age2

-0.033

0.000

-0.008

-0.018

-0.770

0.000

0.269

0.365

Hhsize

0.038

0.000

-0.090

2.631

0.344

0.002

-0.488

-0.572

Children

0.023

0.011

-0.071

0.058

-3.495

0.000

-0.016

-0.015

not

Babies

-0.051

0.000

0.034

-0.055

1.268

0.000

1.259

1.292

Babyf

-0.127

0.000

0.061

-0.122

1.882

0.000

2.694

2.837

HH income /1000

0.032

0.000

-0.005

0.025

0.091

0.250

-0.484

-0.547

Female

-0.375

0.000

-0.121

-0.226

-31.829

0.000

-7.888

-12.383

rural

-0.048

0.004

-0.016

-0.025

-1.701

0.000

-0.611

-0.957

Eduf

0.620

0.000

0.240

0.229

1.179

0.499

Edum

0.460

0.000

0.171

0.180

0.895

0.258

Enrolment ( ) Constant

-2.899

0.000

-8.2341

0.000

-140.342

0.000

Total observations: 27512, sigma error: 0.052 (p-value:0.000); covariance errors:-0.091 (p-value 0.00). Dependent variable: weekly hours worked/100. First column indicates the parameter corresponding to each regressor, the second the p-value. The third and fourth column refers to the hours equation and show the marginal effect of each regressor conditioned to enrolment=1 and enrolment=0 respectively. The standard errors of the marginal effects (col a, b, c, d) are not reported as all the marginal effects were found significant at 5% level.

27

UCW WORKING PAPER SERIES, OCTOBER 2001

Table 8. ML estimates of enrolment and hours worked. Nicaragua Nicaragua enrolment

hours (a)

Regressors

parameter

p value

Tot

(b)

effect| tot

working

(c)

effect| parameter

p value

not working

Tot

(d)

effect| Tot

enrolled

effect|

not enrolled

Age

0.788

0.000

0.465

0.193

6.479

0.000

4.951

age2

-0.041

0.000

-0.012

-0.010

-0.235

0.000

-0.108

-2.328 0.157

Hhsize

0.025

0.061

-0.067

0.005

-0.462

0.001

-0.811

-0.300

Children

-0.049

0.035

0.071

-0.010

0.413

0.091

0.875

0.423

Babies

-0.136

0.000

0.146

-0.030

0.734

0.031

1.835

1.036

Babyf

-0.068

0.096

-0.041

-0.017

-0.568

0.185

-0.439

0.198

HH income /1000000

17.980

0.001

0.329

4.242

63.700

0.140

-9.138

-81.890

Female

0.222

0.001

0.013

0.051

-5.218

0.000

-0.819

-1.090

Rural

-0.443

0.000

-0.144

-0.146

1.300

0.075

0.437

0.617

Eduf primary

0.267

0.000

0.096

0.077

-0.111

-0.145 -0.254

Eduf secondary

0.486

0.000

0.169

0.130

-0.202

Edum primary

0.346

0.000

0.123

0.097

-0.144

-0.185

Edum secondary

0.578

0.000

0.199

0.148

-0.241

-0.297

Constant

-2.914

0.000

Enrolment ( )

-44.786

0.000

-4.120

0.000

Total observations: 4278 sigma error: 66.703 (p-value:0.000); covariance errors: 1.773 (p-value 0.00). Dependent variable: daily hours worked. First column indicates the parameter corresponding to each regressor, the second the p-value. The third and fourth column refers to the hours equation and show the marginal effect of each regressor conditioned to enrolment=1 and enrolment=0 respectively. The standard errors of the marginal effects (col a, b, c, d) are not reported as all the marginal effects were found significant at 5% level with the exception of expenditure and babyf in the hours equation for those who go to school and expenditure, babyf and female in the enrolment equation for those who work.

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

28 

Table 9. Probit Estimates of failing rate at school. Pakistan probability of failing at school

Hours predicted

0.044

0.105

(3.58)**

(1.83)*

Hours>2.5 &4

-0.031 (-0.53)

Hours* trade & manifact

0.059 (3.72)**

Hours*agricolture constr & community

0.042 (3.12)**

Incnet

Hhsize

Children

Female

Rural

Educ father

Educ mother

Constant

Observations

-0.026

-0.027

-0.029

(-2.73)**

(-2.73)**

(-2.95)**

0.031

0.031

0.022

(3.25)**

(3.20)**

(2.20)*

-0.017

-0.016

0.02

(-0.78)

(-0.71)

(0.85)

0.141

0.154

0.105

(2.61)**

(2.85)**

(1.79)*

-0.079

-0.09

-0.06

(-1.85)*

(-2.01)*

(-1.4)

-0.176

-0.182

-0.218

(-4.02)**

(-4.14)**

(-4.86)**

-0.13

-0.129

-0.145

(-1.83)*

(-1.83)*

(-2.04)*

0.313

0.309

0.292

(4.36)**

(4.28)**

(4.03)**

4195

4195

4195

Absolute value of z-statistics in parentheses * significant at 5% level; ** significant at 1% level

29

UCW WORKING PAPER SERIES, OCTOBER 2001

Fig 1 daily hours worked 0 1 2 3 4 5 6 7 8 9 10 11 12

Probability of failing at school

Prob of failing 0.667 0.682 0.698 0.713 0.728 0.742 0.756 0.769 0.782 0.795 0.807 0.819 0.830

0.85 0.80 0.75 0.70 0.65 0.60 0

1

2

3

4

5

6

7

8

9

10

11

12

daily hours worked Fig 2

daily hours worked 0 1 2 3 4 5 6 7 8 9 10 11 12

trade & manif 0.666 0.687 0.707 0.727 0.747 0.765 0.783 0.800 0.816 0.831 0.846 0.859 0.872

daily hours worked 0 1 2 3 4 5 6 7 8 9 10 11 12

Prob of failing 0.675 0.711 0.746 0.763 0.764 0.785 0.806 0.825 0.843 0.859 0.874 0.888 0.901

Probability of failing at school

agric, construction & community

0.666 0.681 0.696 0.710 0.725 0.739 0.752 0.766 0.778 0.791 0.803 0.814 0.825

0.90 0.85 0.80 0.75 0.70

trade & manif

0.65

agric, construction & community

0.60

0

1

2

3

4

5

6

7

8

9

10

11

12

10

11

daily hours worked Fig 3 Probability of failing at school 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0

1

2

3

4

5

6

7

8

daily hours worked

9

12

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

30 

Table 10. Probit estimates of failing rate at school. Nicaragua probability of failing at school. hours predicted

0.029 (3.26)**

quintile 2

0.001 (0.05)

quintile 3

-0.055 (-2.03)*

quintile 4

-0.08 (-2.67)**

quintile 5

-0.123 (-3.41)**

Hhsize

0.004 (-0.97)

Children

0.044 Female

(5.42)** -0.063

rural

(-3.94)** 0.049

edu father: primary

(2.80)** -0.069

edu father: more than primary

(-3.23)** -0.173

edu mother: primary

(-5.90)** -0.085

edu mother: more than primary

(-4.12)** -0.181

Observations Absolute value of z statistics in parentheses * significant at 5%; ** significant at 1%

(

)**

31

hours worked 0 1 2 3 4 5 6 7 8 9 10 11 12

UCW WORKING PAPER SERIES, OCTOBER 2001

probability of Fig 4 failing 0.681 0.717 0.751 0.782 0.811 0.838 0.861 0.883 0.902 0.918 0.933 0.945 0.956

Probability of failing at school

1.00 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0

1

0 0 0 0 0 0 0 0 2 3 4 5 6 7 8 9 10 11 12 0 0 daily hours worked

32 

CHILDREN’S WORKING HOURS, SCHOOL ENROLMENT AND HUMAN CAPITAL ACCUMULATION: EVIDENCE FROM PAKISTAN AND NICARAGUA

REFERENCES Atella, V. and F.C. Rosati (2000), “Uncertainty About Children's Survival and Fertility: A Test Using Indian Microdata”, Journal of Population Economics 13, pp.263-278 Balland, J.M. and A. Robinson (2000), “Is Child Labor Inefficient?”, Journal of Political Economy 108 Basu, K. (1999), “Child Labor: Cause, Consequence and Cure, with Remarks on International Labor Standards” Journal of Economic Literature 37, pp. 1083-1119 Behrman, J., Foster, A.D., Rosenzweig, M.R. and P. Vashishtha (1999), “Women’s schooling, home teching, and economic growth”, Journal of Political Economy 107, pp. 683-714 Becker, G. (1981), “A Treatise on the Family”, Harvard University Press, Cambridge, Mass. Cigno, A., Rosati, F.C. and Z. Tzannatos (2000), “Child Labor, Nutrition and Education in Rural India: an Economic Analysis of Parental Choice and Policy Options”, World Bank W.P. Cigno, A., Rosati, F.C. (2001), “Child Labor, Education, Fertility and Survival in Rural India”, Pacific Economic Review, forthcoming. Grossman, M. (1972), “The Demand for Health: A theoretical and empirical investigation”, NBER, New York. Grossman, M. (1982), “The demand for health after a decade”, Journal of Health Economics, 1, 1 3. Heady, C. (2000) “What is the effect of child labour on learning achievement? Evidence form Ghana”, UNICEF, Innocenti Working Paper, n.79. Maddala, G. S. (1993) “Limited-dependent and qualitative variables in econometrics”, Cambridge University Press. Patrinos H.A., Psacharopulos G. (1994), “Educational Performance and child labour in Paraguay”, International Journal of Educational Development, 15, pp. 47-60

33

UCW WORKING PAPER SERIES, OCTOBER 2001

Psacharopoulos, G. (1997), “Child Labor Versus Educational Attainment: Some Evidence from Latin America”, Journal of Population Economics 10, pp. 377-386 Ravallion, M. and Q. Wodon (2000), “Does Child Labor Displace Schooling: Evidence from Behavioural Responses to an Enrolment Subsidy”, Economic Journal 110, pp. 158-176 Ray, R. (2000), “Child labor, child schooling and their interaction with adult labor: empirical evidence from Peru and Pakistan”, The World Bank Economic Review, 14, n. 2 pp. 347-367 Rosati, F.C. (1999), “Child Labor”. World Bank WP Rosati, F.C. (2000), “Morocco: Reducing Child Labor by Increasing School Availability”. World Bank WP Rosati, F.C. and Z. Tzannatos (2000), “Child Labor in Vietnam: An Economic Analysis”, World Bank WP